The research team analyzed data from 919 women with a body mass index of at least 25 who initiated care through a specialized women’s health telehealth platform between January and December 2024, with follow-up extending through June 2025. More than 60 percent of the cohort carried a diagnosis of the syndrome, identified through electronic health record codes assigned according to Rotterdam criteria, the standard diagnostic framework for the condition. Participants were required to have at least six months of follow-up data for at least one laboratory measure, yielding a mean observation period of 9.6 months during which women attended an average of 6.6 physician visits and 5.8 registered dietitian visits.

The baseline metabolic picture starkly separated the two groups. Women with the syndrome were younger, on average 33.5 versus 36.3 years, but carried a higher body mass index of 35.4 versus 33.4 and weighed more, averaging 210 pounds compared with 198.6 pounds. More striking were the laboratory differences: half of the women with the syndrome showed elevated homeostasis model assessment of insulin resistance, or HOMA-IR, above 4, compared with about a third of women without the diagnosis. Nearly 30 percent had fasting insulin above 24 micro-international units per milliliter, versus 15 percent of their counterparts, and more than a third had elevated ALT, a liver enzyme reflecting hepatic dysfunction, compared with under 19 percent. Low HDL cholesterol and elevated triglycerides were also significantly more common, a pattern consistent with insulin resistance as the central physiological driver of the disorder.

Against this backdrop, the results were remarkable. Across the full cohort, women achieved mean reductions of 14 percent in HOMA-IR, 15.6 percent in fasting insulin, 13 percent in fasting glucose, 6.2 percent in hemoglobin A1c, 21.2 percent in ALT, 25.5 percent in AST, 8 percent in total cholesterol and 4.9 percent in body weight. Every one of these changes was statistically significant after correction for multiple comparisons, both in the overall cohort and separately within each diagnostic group. Crucially, unadjusted comparisons between women with and without the syndrome showed no significant differences in the magnitude of improvement across any outcome, suggesting that the benefits of the care model extended fully to those with the most complex metabolic profiles.

To understand which components of care drove these changes, the researchers used linear regression models adjusted for baseline values, age, body mass index, race and ethnicity, follow-up time and medication status. GLP-1 receptor agonist therapy emerged as the strongest predictor of weight loss, associated with an additional 10.1 pounds of weight reduction when prescribed alone and 7.1 pounds when combined with metformin, compared with no pharmacotherapy. Metformin alone was not associated with significant weight change, though all medication groups showed greater reductions in hemoglobin A1c than women receiving no drugs. Notably, medication status was not significantly associated with changes in insulin resistance markers, fasting glucose, liver enzymes or cholesterol after adjustment, hinting that the care model itself, rather than pharmacology alone, may be responsible for much of the broader metabolic improvement.

That interpretation gains support from the outcomes among women who received no medications at all. Roughly 12 percent of the cohort fell into this category, and even they achieved significant reductions in weight, fasting glucose, liver enzymes and total cholesterol. Because access to GLP-1 receptor agonists remains constrained by cost, supply and side effects, this finding carries practical weight: structured behavioral and dietary support delivered through a telehealth model appears capable of moving metabolic markers independently of drug therapy. The authors suggest this supports the value of behavior change and dietary optimization alongside, or in place of, pharmacological management.

The study also addressed a question of direct clinical relevance: whether GLP-1 receptor agonists work as well for women with the syndrome as for those without. The absence of a significant interaction between diagnosis and medication group for weight loss indicates that the weight-reducing benefit of these drugs generalizes across both populations, an important consideration given the high prevalence of overweight and obesity among affected women. One exploratory signal did emerge for hemoglobin A1c, where women with the syndrome who received no medication saw their A1c rise while their counterparts without the diagnosis saw it fall. The authors caution that this interaction, initially significant at a p value of 0.011, did not survive Bonferroni correction across the eight interaction tests performed, and should therefore be interpreted cautiously as potentially reflecting chance.

The magnitude of improvement deserves context. Reductions of 14 to 16 percent in insulin resistance markers and more than 20 percent in liver enzymes over roughly nine months compare favorably with outcomes reported in structured in-person intervention programs for these patients, which typically demand far greater resources and travel. The weight loss observed with GLP-1 receptor agonists also aligns with results from large-scale telehealth anti-obesity medication programs, reinforcing that virtual care can reproduce clinical trial-adjacent results outside controlled research settings. What distinguishes the current analysis is its multi-marker approach: rather than focusing solely on weight, it captured insulin dynamics, glycemic control, hepatic function and lipid profile, dimensions particularly relevant for a population in which insulin resistance and liver dysfunction may persist or predominate independently of weight change.

The authors are careful to enumerate limitations. The retrospective design precludes causal inference, medication assignment was not randomized, and prescribing patterns likely reflected differences in clinical severity and patient preference. Diagnosis was inferred from billing codes, introducing possible misclassification in both directions, and prescription records may not perfectly reflect actual medication use, biases that would tend to push results toward the null. The cohort was drawn from a single platform serving a predominantly privately insured population, limiting generalizability, and the requirement for at least six months of follow-up excluded women who disengaged early, who tended to have less severe metabolic profiles. Without a control group, causation cannot be firmly established, though the authors note that longitudinal data show untreated cohorts typically worsen over time, lending indirect support to a genuine treatment effect.

Despite these caveats, the findings arrive at a moment when access to integrated metabolic care remains a persistent barrier for millions of women. Recent clinical guidance emphasizes the importance of multidisciplinary management addressing both metabolic and reproductive dimensions of the syndrome, yet fragmented care pathways and shortages of specialists with relevant expertise continue to leave many patients undertreated. This study suggests that a telehealth model pairing physician oversight with registered dietitian support can deliver metabolic outcomes comparable to resource-intensive in-person programs, even for women at elevated metabolic risk. As virtual care platforms scale and payers weigh coverage decisions, real-world evidence of this kind may prove instrumental in determining whether structured metabolic care becomes a routine standard for a condition long dismissed as merely reproductive.

Subject of Research: Real-world metabolic outcomes of multidisciplinary telehealth care in women with and without polycystic ovary syndrome

Article Title: Real‐World Metabolic Improvements in Women With and Without Polyendocrine Metabolic Syndrome via a Multidisciplinary Telehealth Care Platform

Article References: Huddleston, H. G., Blank, R., Sommers, N., Hellman, N., Ramallo, F., Pham, T., & Lindquist, K. (2026). Real‐World Metabolic Improvements in Women With and Without Polyendocrine Metabolic Syndrome via a Multidisciplinary Telehealth Care Platform. Endocrinology, Diabetes & Metabolism, 9(6), Article e70332. https://doi.org/10.1002/edm2.70332

Image Credits: AI Generated

DOI: 10.1002/edm2.70332

Keywords: PCOS, telehealth, insulin resistance, metabolic health, GLP-1 receptor agonists, metformin, HOMA-IR, liver enzymes, HbA1c, registered dietitian, women's health, obesity